In 2018, Canada revamped a long-existent but rarely used prohibition of fake news in the run-up to an election. According to the Canada Elections Act, making or publishing false statements about political figures’ citizenship, place of birth, education, membership in a group, legal offenses or professional qualifications—if done with the intention of influencing the results of an election—could result in a fine of $50,000 or 5 years of prison.
A few months ago, following a survey of Canadian case law, I wrote here that any legal challenge to this provision was unlikely to have much success. Perhaps I should consider throwing my crystal ball away: On Feb. 19, Ontario’s Superior Court of Justice declared that Section 91(1) of the Canada Elections Act—the section of the law on making or publishing false statements about political figures—infringes on freedom of expression protected by Section 2b) of the Canadian Charter of Rights and Freedom and was therefore of no force or effect.
Let’s take a look at what happened.
The case was brought by the Canadian Constitution Foundation (CCF), a registered charity that defends the rights and freedoms of Canadians through litigation. PEN Canada, a literary association fighting for freedom of expression, also intervened in support of the CCF.
CCF asked for Section 91(1) of the Canada Elections Act to be declared invalid because it infringes on freedom of expression in a manner unjustifiable in a free and democratic society. While acknowledging that the provision was indeed a restriction of an expressive activity protected by the Charter, the attorney general of Canada argued that it met the criteria to be considered a justified limit under the constitution and should therefore be preserved.
This set the stage for what could have been an interesting debate over how to strike a balance between the free circulation of political ideas that is so important to democracy and the preservation of public confidence in democratic institutions, which requires the distribution of false information to be contained.
This debate seems to have taken place in court—the petitioners questioned the very premise that Section 91(1) achieves salutary effects and suggested less restrictive measures instead, for instance—but that argument was not part of the judge’s reasoning. Instead, the case was ruled on a narrow question of statutory interpretation.
The question at hand was to decide whether or not the actor spreading false statements was required to have any knowledge of their falsity to be prosecuted under Section 91(1) of the Canada Elections Act. In other words, the issue hinged on whether or not the dissemination of accidental and unknown falsehood was prohibited.
The attorney general of Canada conceded that if accidental or unknown falsehoods were—as claimed by the petitioners—covered by Section 91(1), the provision would not represent a justifiable limit on freedom of expression. Thus, the only question was to determine how to interpret the law.
The provision reads as follows:
91 (1) No person or entity shall, with the intention of affecting the results of an election, make or publish, during the election period,
(a) a false statement that a candidate, a prospective candidate, the leader of a political party or a public figure associated with a political party has committed an offence under an Act of Parliament or a regulation made under such an Act—or under an Act of the legislature of a province or a regulation made under such an Act—or has been charged with or is under investigation for such an offence; or
(b) a false statement about the citizenship, place of birth, education, professional qualifications or membership in a group or association of a candidate, a prospective candidate, the leader of a political party or a public figure associated with a political party.
This provision has existed, in one form or another, since 1908, and for most of its life, it expressly required that the person or entity making the false statement did it with the knowledge that the information was false.
In 2018, following growing concerns over disinformation, the provision was rewritten. In this overhaul that expanded the provision’s scope and aimed to clarify it, the word “knowingly” disappeared.
For the attorney general, the removal of the word “knowingly” was nothing more than a housekeeping change designed to prevent redundancy in the provision: The knowledge requirement still exists implicitly.
This argument did not convince the court, which engages in an interpretative exercise of the provision following Elmer Driedger’s principles: “[T]he words of an Act are to be read in their entire context and in their grammatical and ordinary sense harmoniously with the scheme of the Act, the object of the Act, and the intention of Parliament.”
From a textual analysis standpoint, the provision does not contain any wording indicating that an actor must have been aware that the statement was false to be prosecuted under the provision. This becomes important when the provision is read in the context of the whole statute. In other similar provisions of the Canada Elections Act, the law explicitly states that a proof of knowledge is required. Section 91(1) doesn’t, and according to the judge, this silence speaks volumes.
To understand Parliament’s intention behind removing the word “knowingly,” contextual arguments such as comments made by politicians and government officials could have been useful to show that the government’s intent was genuinely not to change the meaning of the provision. Had they been persuasive, these arguments could have countered the textual arguments.
However, according to the judge, the only argument brought by the attorney general that could have explained the absence of the word “knowingly” in Section 91(1) was legally inaccurate and therefore “cannot be taken as reflecting Parliament’s true intention.”
The testimony of the director of investigations with the Office of the Commissioner of Canada Elections—which submitted an affidavit explaining that despite the changes in the law, the commissioner still considered a proof of knowledge to be required to move forward with prosecution—did not succeed either in convincing the court that the word “knowingly” was implicit in the text of Section 91(1) even though the word wasn’t written in the text.
In fact, the affidavit might have even done the opposite. By suggesting that, according to the commissioner’s interpretation, a person or entity who was “reckless about the truthfulness of the statement” could be prosecuted under Section 91(1), the Office of the Commissioner of Canada Elections put the final nail in the attorney general’s claim that knowledge of spreading falsehoods was an implicit requirement to be prosecuted under 91(1).
Indeed, according to Canadian jurisprudence, the recklessness threshold requires that someone acts despite knowing the risks, whereas a proof of knowledge is necessary to meet the “knowingly” requirement. Therefore, the commissioner’s interpretation under which someone could be prosecuted for reckless conduct would not be possible if the word “knowingly” was—implicitly or explicitly—stated in Section 91(1).
The court therefore concluded that accidental or unknown falsehoods were prosecutable under the current form of Section 91(1), a result that—as conceded by the attorney general—would pose an unjustifiable limit to freedom of expression. Consequently, 91(1) has been declared of no force or effect and the Canadian government ought to go back to the drawing board pretty soon if it wants the provision reenacted before the next elections, which could be declared at any time in the coming year.
Contrary to what the judgment could lead us to believe, the task might require much more than simply adding the word “knowingly” somewhere in Section 91(1). Many questions brought in this lawsuit were left unanswered and will require further reflection. To be constitutional, should the protection against false statements apply only to political candidates, prospective candidates and party leaders and not to public figures associated with a political party? Is the terminology used—words like “committed an offence” for instance—too vague, and should it be more precisely defined? Should Section 91(1) provide for an exception regarding satire similar to the one existing for impersonation of political candidates under Section 480.1(1)?
And, maybe more importantly, knowing that in more than a hundred years of the provision’s existence, nobody has ever been convicted of the offense provided by Section 91(1) or its predecessor, is such a prohibition really the best tool to address the disinformation problem?
It’s an important question, but it’s a tough one. And it’s not unique to Canada. The European Union and the United States, for instance, are also pondering the best measures in the legislative toolbox to tackle disinformation.
Author: Eve Gaumond
This post first appeared in Lawfare. Read the original article.